Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.openai-nebula.com/llms.txt

Use this file to discover all available pages before exploring further.

​
Overview

Codex is primarily designed as a terminal-based coding agent. Using Nebula API (which is OpenAI-compatible), you can access popular GPT models like gpt-5.4 and gpt-5.2 through a single API key.

​
Prerequisites

  • Create an API key in the Nebula Api console
  • Install Codex on your machine (follow Codex installation instructions for your OS)

​
Configure Nebula Api

Codex requires an OpenAI-compatible API key and Base URL.
Use https://llm.ai-nebula.com/v1 as the Base URL (must end with /v1).

​
Configuration (TOML - required)

Environment variables may not work reliably. Please configure Nebula directly in your TOML config. Define a custom provider and point it to the /v1 base URL: 
model = "gpt-5.4"
model_provider = "openai-custom"
personality = "pragmatic"
model_reasoning_effort = "high"

[model_providers.openai-custom]
name = "Nebula"
base_url = "https://llm.ai-nebula.com/v1"
wire_api = "responses"

# If your Codex config does not automatically inject the API key for custom providers:
[model_providers.openai-custom.http_headers]
Authorization = "Bearer YOUR_API_KEY"
Content-Type = "application/json"

​
Choose a model

Pick a GPT-family model ID from the Nebula model list.
Codex currently supports GPT-family model IDs only (for example gpt-5.4).

​
Personality & Performance

Codex allows you to customize how the model behaves using two optional settings: personality and reasoning effort.

​
Personality

The personality setting controls the tone and style of responses. 
personality = "pragmatic"
Common options:
  • pragmatic → direct and practical, minimal explanation
  • concise → short and to the point
  • detailed → more explanation and context
  • analytical → structured and logic-focused
  • creative → more expressive and flexible
Personality is a soft preference and may not always be strictly enforced. For precise control, use a system prompt.

​
Reasoning Effort (Performance)

The model_reasoning_effort setting controls how much effort the model uses when solving problems. 
model_reasoning_effort = "high"
Available options:
  • low → fastest responses, simpler reasoning
  • medium → balanced speed and quality
  • high → deeper reasoning, better for complex tasks
Higher reasoning effort may increase response time but improves accuracy for coding, debugging, and complex logic.

​
Example

model = "gpt-5.4"
model_provider = "openai-custom"
personality = "pragmatic"
model_reasoning_effort = "high"
This configuration provides:
  • clear and practical responses
  • stronger reasoning for complex tasks

​
Troubleshooting

  • 401 / invalid API key: re-check your key in the Nebula console and make sure you didn’t add extra spaces
  • 404 / wrong endpoint: confirm the Base URL ends with /v1
  • Model not found: verify the model ID exists in the Nebula model list